Risk and Protective Factors in Relation to Early Mortality Among People With Serious Mental Illness: Perspectives of Peer Support Specialists and Service Users

Abstract

Objective:

Individuals with serious mental illness (SMI) experience a 10–25-year reduced life expectancy when compared to the general population that is due, in part, to poor health behaviors. Yet, in spite of the development of health promotion and self-management interventions designed for people with SMI to promote health behavior change, the mortality gap has increased, suggesting that relevant factors are not being addressed. The objective of the present study was to explore potential contributors to early mortality among individuals with SMI by drawing from the lived experience of certified peer support specialists and service users (SUs).

Method:

Face-to-face semistructured interviews were conducted with a convenience sample of SU participants (n = 17) and certified peer specialists (n = 15). Qualitative data were analyzed using a grounded-theory approach.

Results:

We identified a final set of 27 codes relating to five overarching themes that relate to both risk factors and protective factors for early death: social connectedness (24.1% of coded items), treatment (21.3%), coping (21.3%), physical health and wellness (18.5%), and resilience and mental health (14.8%).

Conclusions and Implications for Practice:

Findings add to the literature supporting the powerful role of social processes in shaping health in people with SMI beyond social determinants of health (SDOH; e.g., income, employment) and health behavior change. Interventions that reduce loneliness and isolation in combination with addressing more conventional SDOH may have the most potential to reduce early mortality in people with SMI.

Individuals with serious mental illness (SMI), such as bipolar disorder, schizophrenia, and major depressive disorder, have a reduced life expectancy of approximately 10–25 years compared to the general population (de Hert et al., 2011). Health behaviors including poor diet and sedentary lifestyle cause up to half of preventable early deaths in the U.S. general population (McGinnis & Foege, 1993). Early deaths in people with SMI are also due, in part, to modifiable lifestyle behaviors including substance use, tobacco use, diet, and physical activity (Druss, Chwastiak, et al., 2018; Firth et al., 2019; Liu et al., 2017; Mauer, 2006). These findings led to the development of health promotion and self-management interventions designed for people with SMI, which typically promote behavior change by teaching skills to help people participate in treatment and encourage healthy behaviors (Firth et al., 2019; Naslund et al., 2017; Whiteman et al., 2016). However, the mortality gap between those with SMI and the general population continues to increase (Walker et al., 2015). While likely due to the scalability and implementation issues for health promotion and self-management interventions (Whiteman et al., 2016), this also prompts the question of whether there are factors associated with early mortality among people with SMI that are not addressed or could be more optimally targeted in these interventions, for example, by incorporating support from friends and family (Aschbrenner, Mueser, et al., 2017; Aschbrenner, Naslund et al., 2017).

There is increasing recognition of the influence of social determinants of health (SDOH; i.e., qualities of the environments in which individuals live their day-to-day lives) on a wide range of psychiatric and quality-of-life outcomes. For example, the U.S. Department of Health and Human Services has committed to addressing SDOH such as health care access and social/community contexts in the service of improving people’s lives (Healthy People 2030). Individuals with SMI experience inequities in several SDOH, including stigma, poorer access to care, and economic disadvantages that likely contribute to poor mental health and premature mortality (Alegría et al., 2018). Some SDOH, such as social cohesion, are underacknowledged and understudied, particularly among people with SMI—leading to potential missed opportunities for intervention (Fortuna et al., 2022).

Although individuals with SMI are uniquely positioned to inform understanding of unmet needs related to premature mortality because of their lived experience expertise, there are limited qualitative data exploring their perspectives on this topic. Extant studies of the perspectives of people with SMI on health behaviors more generally have identified barriers including stress, stigma, isolation, lack of social support, challenges within the health care system (Blixen et al., 2016), limited experience with physical activity and its benefits, and the negative impacts of medications and anxiety (Johnstone et al., 2009). At the point of contact for intervention, people with SMI have reported barriers such as mental and physical conditions (Pearsall et al., 2014) and ambivalence about changing unhealthy coping behaviors that they find calming (Kerr et al., 2013; Pearsall et al., 2014). Interventions designed to improve health behaviors have in fact been thought to be effective because of factors such as the provision of support from loved ones (Aschbrenner et al., 2013) and opportunities for practice and learning from others (Muralidharan et al., 2021).

The lived experience of peer specialists, in particular, is an important resource in generating novel strategies for interventions in which people with SMI may be most interested and willing to engage. Certified peer support specialists (PSSs) are trained and accredited by their respective states to use their own experiences with mental health challenges and offer support to support others’ recovery. Qualitative research has revealed that PSSs provide meaningful contributions to health behavior interventions for people with SMI, with disclosure of their lived experience of changing health behaviors serving as motivation for intervention participants (Bochicchio et al., 2019). It is essential to learn more about how PSSs could advance the mission of reducing premature mortality among people with SMI.

Given that there are limited qualitative data exploring the perspectives of people with SMI with respect to reasons for early mortality in this population, the objective of the present study was to fill this gap in the literature. We used an exploratory qualitative approach to identify potential contributors to early mortality among individuals with SMI by drawing from the lived experience of mental health service users (SUs) and certified PSSs to order to inform the advancement of theory about risk and resilience factors in mortality among people with SMI.

Method

Participants and Study Design

Face-to-face semistructured interviews were conducted with a convenience sample of SU participants (n = 17) and certified PSSs (n = 15) for the purpose of the parent study focused on barriers and facilitators of participation in a digital peer support intervention. Eligibility criteria for SUs were as follows: (a) age 18 or older with a documented diagnosis of schizophrenia, schizoaffective disorder, bipolar disorder, or persistent major depressive disorder based on the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (American Psychiatric Association, 2013); (b) enrollment in mental health treatment for at least 3 months; (c) a diagnosis of cardiovascular disease, obesity, diabetes, or chronic obstructive pulmonary disease in the medical chart; (d) ability to speak and read English; and (e) ability to provide voluntary informed consent for participation in the study.

PSSs in this study met criteria to be a Massachusetts certified peer specialists. Requirements included self-report of any mental health diagnosis, engagement in active treatment, completion of an 80-hr training that includes coursework and group activities focused on the fundamentals of peer support, competence in partnering of peers and SUs with an understanding of cultural differences and human experience (i.e., nonclinical, person-first) language, and passing a written examination. They also needed to be able to speak and read English and provide voluntary informed consent for participation in the study.

All participants were recruited from an urban community mental health center in the United States via clinician referral. After written informed consent was obtained, participants completed 30–60-min face-to-face interviews without others present in a familiar clinical setting; interviews were audio-recorded and transcribed. Interviews were conducted by three trained interviewers with interest in understanding early mortality among individuals with SMI. Participants were paid $20 in cash in-person directly following the interview. All procedures were conducted in accordance with the ethical standards of the institutional review board at Dartmouth and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.

The semistructured interview guide for the parent study was codeveloped with two certified PSSs using the Peer and Academic Model of Community Engagement (Fortuna, Brooks, et al., 2019). SUs and certified PSSs were asked the same questions. The present study leveraged these interviews to explore potential contributors to early mortality among individuals with SMI by focusing on participant responses to two questions: (a) “Why do you think people with a lived experience of a mental health condition have a shorter lifespan than people who don’t have a mental health condition?” and (b) “What types of interventions could help people with SMI live longer?” Probing questions referenced SDOH and other dimensions of health such as loneliness; stigma; lack of community inclusion; health behaviors; system issues—health insurance, costs, access; and medication. Qualitative interviews were conducted until saturation of data related to the original research question about engagement in digital peer support interventions was detected.

Data Analysis

Transcripts for the 32 participants (17 SUs, 15 certified PSSs) were analyzed using a grounded-theory approach (Martin & Turner, 1986). As such, we did not develop a priori codes, theories, or hypotheses prior to coding; rather, codes were developed organically via review of the data. When data were applicable to more than one domain, we assigned codes and themes using the “best fit” approach. We assessed and resolved within-group disagreements using a “member-checking” approach (Birt et al., 2016). The research team contacted four of the participating PSSs to validate key themes that emerged from the data and resolve discrepancies. The second and third authors reviewed the transcribed data and assigned initial codes and operational definitions, a validated approach that allows for multiple perspectives (Martin & Turner, 1986).

The first author used thematic analysis to identify patterns in the data-derived codes, which were translated to overarching themes (Braun & Clarke, 2006). The search for, identification, and categorization of themes into risk and protective factors were informed by paradigms of risk and resilience in mental health (Southwick et al., 2014). Themes were defined and named in a recursive, iterative manner in which it was ensured that each theme was coherent, adequately distinct from other themes, and had enough data to support it. Final themes were reviewed and agreed upon by the research team in light of how they apply to people with SMI, ensuring that the themes addressed all of the coded data (Kiger & Varpio, 2020). Thematic coverage of coded data and how well themes address the question of interest in this study was again confirmed by the team during preparation of the figure summarizing themes, which was initially prepared by a co-author unfamiliar with the initial coding (Figure 1). We also triangulated data from SUs and certified PSSs and noted similarities and differences. In sum, thematic analysis was used to organize and interpret the data in order to address the goals of the study.

Figure 1.

Summary of Key Themes and Codes for Risk and Protective Factors

* Number of times that the theme and code were reflected in the data. ** Percentage of total coded items (N = 108).

Results

Demographic Characteristics

As displayed in Table 1, SU participants (n = 17) were predominantly male (70.6%) and White (82.4%), with a mean age of 51.2 years (SD = 8.8, range = 38–75). Primary psychiatric diagnoses of SUs were major depressive disorder (29.4%), schizophrenia spectrum disorders (41.1%), and bipolar disorder (23.5%). PSSs (66.7% female, 86.7% White) were on average 39.7 years of age (SD = 12.1, range = 24–61 years). PSSs were not required to disclose their mental health diagnoses as they are employees and protected by the American with Disabilities Act. All PSSs were currently employed in the field of mental health.

Table 1.

Demographic Characteristics of Participants

	Service users (n = 17)		Certified peer specialists (n = 15)
Variable	n	%	n	%
Gender
Male	12	70.6	5	33.3
Female	5	29.4	10	66.7
Age
20–39	2	11.8	6	40
40–59	14	82.4	4	26.7
>60	1	5.9	2	13.3
Race/ethnicity
White	14	82.4	13	86.7
Black/African American	1	5.9	1	6.7
American Indian/Alaskan Native	—	—	—	—
Asian	—	—	—	—
Native Hawaiian/Pacific Islander	—	—	—	—
Multiracial	2	11.8	1	6.7
Education
Some elementary schooling	1	5.9	—	—
Completed 8th grade	1	5.9	—	—
Some high school	5	29.4	—	—
Completed high school or GED	6	35.3	—	—
Some college	2	11.8	6	40
Completed college or technical school	1	5.9	—	—
Completed associate’s degree	1	5.9	1	6.7
Completed bachelor’s degree	—	—	5	33.3
Some graduate school	—	—	2	13.3
Completed master’s	—	—	1	6.7
Psychiatric diagnoses
degree Bipolar disorder	4	23.5
Schizophrenia	4	23.5
Schizoaffective disorder	3	17.6
Major depressive disorder	5	29.4

Note. GED = General Educational Development.

Themes

We derived a final set of 27 codes relating to five themes: social connectedness, treatment, coping, physical health and wellness, and resilience and mental health. Each theme included between 1 and 5 codes or subthemes that were conceptualized as either a protective or risk factor, summarized in Figure 1. See Table 2 for selected quotes illustrating each theme.

Table 2.

Representative Quotes for Each Theme

Theme	Risk factors: sample quote for most frequent code	Protective factors: sample quote for most frequent code
Social connectedness	Loneliness: And their biggest fear is being alone. And one point they don’t want someone being next to them, bothering them “cause they feel like they’re followed or somebody’s gonna hurt them. Or, they feel left out-I mean really left out in the cold because when like birthdays, Christmas rolls around,” important events in their life, uh they feel no one’s gonna bother. (SU)	Social support: Uh, definitely more support um, like emotional peer support, family support. It seems that a common theme with people with a mental illness is they feel loneliness. And so, by making some type of connection, you know preferably more than one, that they can talk to on a regular basis that is supportive. I think that makes a difference. (PSS)
Coping	Substance use: They die because they take illegal drugs to self-medicate. Um, I mean I’ve got friends that are proud of their vaping and pot heads and everything else and I’m thinking are you crazy? Because they’re ruining a perfectly good body. I want to strangle them. Um, but there’s nothing you can do. You can only do what’s best for you and hope your friends do what’s best for them. (SU)	Activities: We try to find something that the-the client enjoys, whether it’s painting, music, and try to find something geared to that person’s, you know- Um, to that person’s liking so we can get him so he’s not sitting in his apartment all day long doing nothing. (PSS)
Treatment	Medication side effects: Drugs. I’m on sixteen different medications. I’m talking about prescribed legal medications. I’m on 3 for blood pressure. I’m on 2 for the shaking. I’m on the Latuda for the uh bipolar and I could go on and on. And yeah, it gets to be a drag and it’s bad for your liver. But you have, it might be a shortened life but if it’s better quality. (SU)	Effective mental health care: And if we have more uh, say technology with, say psychology- good psychology and therapists who really care-And clubhouses together, there’d be less hospitalizations. (SU)
Physical health and wellness	Poor health behaviors: Um, smoking, overeating, um not having motivation or desire, therefore neglecting your medical health or your mental health needs. (SU)	Physical activity: Going to nutrition classes, better healthcare, better exercise, taking care of your body better. (SU)
Resilience and mental health	Suicide: It’s lack of hope and prospects and being like, you know worst case scenario they kill themselves. (SU)	Purpose in life: Maybe (a place) with animals, maybe, and they, you know, maybe people could volunteer and like feel like they’re needed. That’s a big one too, feeling that you’re needed, have a purpose in life. (SU)

Note. PSS = peer support specialist; SU = service user.

Social Connectedness

Social connectedness was the most prevalent theme (24.1% of coded items), with 14 statements about the value of the presence of social connectedness (i.e., social support and community integration) and 12 statements pertaining to the contribution of social isolation and loneliness to early mortality. In terms of risk, respondents described negative consequences of being isolated from loved ones, wishing for stronger family and peer support, and fears of being alone and excluded from social events. The seriousness of the consequences of loneliness was evident via statements such as “I’ve said it like fifteen times, loneliness” (PSS); “their biggest fear is being alone” (SU); and “Not having family support is a big one for me” (SU). A PSS described some of the functions of social support: “Having no family or friends that care about you in the right way to boost your self-esteem and your feelings about just having a place in the world.” PSSs described risks of loneliness relative to health problems: “Loneliness is more dangerous than smoking”; “loneliness kills you more than heart disease”; and how it intersects with poor engagement in care: “I think people without supports just don’t care. Maybe they might not run to the doctors’ offices.” Social connectedness often intersected with other themes, such as engagement in suicide prevention (PSS: “More communication lines open. People who are contemplating suicide really need people to talk to”) and health behaviors:

SU: Due to the pressure you just like, you become kind of lazy. You don’t want to walk or exercise. I think that if the people with mental illnesses, including myself would have something to hold them together and maybe do it as a group.

Several SUs described the value of clubhouses and community centers to meet social needs (e.g., “Having a close-knit community is, [organization name] clubhouse. [Organization name] clubhouse is literally, not just a community of friends and acquaintances, we’re family.”) Some statements also intersected with relaxation and purpose in life:

Communicating, socializing, I still come here to socialize. I like the music, I like the people, I like the atmosphere;

I think it really comes from the person within, like in their, you know, ability to connect with other people and have a reason to keep moving forward. Like, I have my kids, not everyone has kids. My kids are the reason I’m alive;

and I had no idea what psych rehab was, and I went to this interview in [organization name] and I was like man, community integration, that sounds great-I like doing things for the community.

Coping

Several respondents highlighted the high levels of stress experienced by people with SMI and identified strategies for coping with stress that are relevant to early mortality (21.3% of coded items). PSSs emphasized the “survival mode” that many people with SMI experience: “I feel like they’re trying so hard to survive their life, but the things in which they have to do to survive that life because it’s so painful at times, is very consequential to longevity”; and

I think the chronic continuation and following that mental health takes on a person and what people do to survive is such an exhausting process. I mean, people who are mentally ill are resilient people because even if they fall into patterns of addiction or harming, it’s survival mode for them.

Specific maladaptive coping strategies included general risky behaviors and specifically substance use (e.g., alcohol, illicit drugs). PSSs described the challenges of healthy coping in the context of high levels of stress: “It’s a lot easier and more pleasurable to have a lot of unhealthy coping skills than it is to have healthy coping skills” and “Eating and drinking to numb feelings and pain is a lot easier than actually putting in work and going to the gym and working out and you know, doing other things.” General relaxation and specific strategies such as painting and music were identified as healthy modes of coping. For example, a PSS stated, “We try to find something that the client enjoys, whether it’s painting, music, and try to find something geared to that person.”

Treatment

Respondents described the benefits of engaging in various forms of mental health treatment, as well as negative consequences of treatment and the lack of treatment (21.3% of coded items). The most frequently identified negative consequence of treatment was side effects of medications: For example, medications are “damaging” and “take a toll on the body,” and can lead to unintended overdose (SU: “it was an overdose of an antipsychotic medication forced upon him by a certain psychiatrist without given sufficient informed consent on it”). Several participants also described experiencing a lack of understanding by providers and feeling stigmatized by them, for example, as “scary” and “med-seeking.” With respect to protective factors, respondents described the value of mental health treatment (both medications and psychotherapy), saying that it is “necessary” and is helped by active participation and asking questions. Intersecting with social connectedness, they also highlighted the value of strong relationships with providers: “I feel like I heard something about like having a like primary care provider (PCP) that you have a good relationship, like a great relationship with could like save you so many years or something like that” (PSS).

Physical Health and Wellness

Participants identified health behaviors as key to early mortality (18.5% of coded items). Regarding risk, respondents identified poor health behaviors such as unhealthy eating, avoiding doctors’ appointments, and chronic health conditions such as diabetes, obesity, and pain as related to early mortality. Intersecting with coping and treatment, an SU said, “I think a lot of people with mental illness have diabetes and are overweight. Or their high stress levels and they don’t eat as healthy. And they tend to avoid the doctor, like doctor appointments.” One PSS said, “physical health. When you don’t feel well mentally that shit goes out the door and we know from research that wellness and physical wellness is helpful to living longer” and “you know, and obesity, diabetes, so important to get those under-you know under- under control.”

Resilience and Mental Health

Several interrelated aspects of resilience and mental health were identified as related to early mortality (14.8%). The most frequently described risk factor was suicidality, with several participants typically stating very simply that many individuals with SMI are at elevated risk of suicide. SUs reflected on their own suicide attempts. With the caveat that participants may have been cued to describe digital interventions given the nature of the overall interview, PSSs identified technology-based interventions as potentially helpful for suicide prevention:

People who are contemplating suicide really need people to talk to. So, an app might be helpful in that sense and to give people more- more than just the suicide lifeline to call. They could contact somebody on the-on the app that might help them with their contemplation.

So, like how can we utilize what we have as resources, as technology, to offer like the alternative viewpoint (to poor self-worth)? Even though a lot of people won’t want to hear it, if you continue to push it in someone’s face, they’ll start to hear it a little bit.

All other subcodes were identified by only one or two people, but included important risk factors for suicide (e.g., depression and hopelessness) and protective factors with respect to suicide (e.g., family support, purpose in life, hope, optimism, self-worth). For example, a PSS stated, “if it wasn’t for my kids, I probably would have like have killed myself on that deployment. Because I almost tried and then I just like realized that they needed me, you know, so I couldn’t do it.” PSSs offered ideas about interventions involving hope:

I think understanding that life is worth it whatever kind of intervention we want to call that, but hope- I hate the word hope. It’s very fluffy, and annoying, and overused, but I think the concept is when there’s so much that life is dealing you that isn’t great, if you don’t have something to cut through that at some point, like you’re not going to survive. None of us are strong enough to have constant doom and gloom without a glimmer here and there. We’ll all just suffer;

and So, I think that employing interventions that have hope, optimism, promote self-worth- A lot of people fall into the darkness of mental health because they truly don’t think they deserve better.

Discussion

In this study, we drew from the lived experience of mental health SUs and certified PSSs to identify potential underaddressed contributors to early mortality among individuals with SMI. Our sample identified several risk and protective factors, with loneliness and social connectedness as the most predominant themes related to early mortality, followed closely by (and often intersecting with) coping strategies, treatment engagement, physical health and wellness, and resilience and mental health.

By incorporating the perspectives of SUs, we extended findings from a recent study of PSSs (Fortuna, Ferron, et al., 2019) which similarly identified loneliness and social isolation as unmet needs among people with SMI. Social relationships are increasingly seen as critical for understanding both physical and mental health (Hawkley & Cacioppo, 2010; Holt-Lunstad, 2018; Holt-Lunstad et al., 2010). Stronger social relationships are associated with a 50% increased likelihood of living longer, an effect that outweighs those of several health behaviors (Holt-Lunstad et al., 2010), perhaps due to feelings of emotional closeness, shared identity and common bonds, valued relationships, involvement and social engagement with others, and feeling cared for and accepted (Hare-Duke et al., 2019).

Just as having strong social relationships is protective, the experience of loneliness (which is twice as common among individuals with SMI than those without; Perese & Wolf, 2005) confers risk for many negative outcomes (Hawkley & Cacioppo, 2010), including accelerated aging (Hawkley & Cacioppo, 2007). Among those with SMI, this may be due in part to loneliness and poor social support being barriers to engaging in health behaviors and self-management (Blixen et al., 2016; Fortuna et al., 2021; Johnstone et al., 2009). People with SMI often have poor social skills that render it more difficult to develop and maintain close and valued relationships. They lack normative opportunities to engage in social activities, and as identified by our sample, experience stigma related to their SMI. People in their social networks who do not have SMI may have difficulty understanding their experiences. Loneliness is argued to elicit the perception of a lack of safety that perpetuates hypervigilance for threat (Hawkley & Cacioppo, 2010), a common issue among people with SMI described as “survival mode.” Each of these potential mechanisms can be targeted in intervention. In doing so, possibly early mortality could be prevented, as well as quality of life and psychiatric symptoms such as delusions (Garety et al., 2001; Horan et al., 2006).

The additional themes observed in this study are highly complementary to each other and to social connectedness and represent common themes identified in studies of barriers to and interventions for improving health behaviors and identifying unmet needs of people with SMI (Fortuna, Ferron, et al., 2019). High levels of stress interfere with engaging in healthy behaviors, including among people with SMI (Blixen et al., 2016; Firth et al., 2016). People with SMI are therefore among the those who may benefit from learning effective coping strategies to take the place of unhealthy but calming behaviors like substance use (Pearsall et al., 2014), with prevention of drug use identified by PSSs as a critical need for people with SMI (Fortuna, Ferron, et al., 2019). Participants in this study identified healthy coping strategies such as engaging in activities and relaxation, which can be shared with and reinforced by members of one’s support system (Aschbrenner et al., 2013; Aschbrenner, Mueser, et al., 2017). Person-centered approaches that include tailoring new coping strategies according to individual preferences, strengths and resources, and contexts (including interpersonal contexts) are critical for ensuring maintenance of strategy use over time since stressors, attributions, and the effectiveness of specific strategies can vary widely—including among people with SMI (Folkman & Moskowitz, 2004). Reducing the additive effects of chronic stress and consequences of unhealthy coping strategies should have a positive impact on mortality.

Several participants noted that engagement in effective mental health treatment is key to preventing early mortality. However, participants in this study also noted that people with SMI often have untreated mental health needs and that, even when engaged in treatment, may have difficulty complying with treatment recommendations; they also identified negative consequences of treatment such as medication side effects and other care-level issues such as lack of understanding by providers (Blixen et al., 2016; Johnstone et al., 2009). These findings reinforce the need for well-trained, sensitive providers to collaboratively develop person-centered treatment plans that accommodate patient preferences and provide information in a clear and understandable manner (to increase health literacy, also identified as a protective factor in this study). Access to services and continuity in relationships with service providers are also aspects that influence if health needs are met (Storm et al., 2019). Shared decision-making should also be revisited throughout a course of treatment to address complications such as side effects of medications that are intended to be helpful but can also interfere with day-to-day living. Honest, open dialogue between patients and providers is critical for ensuring that treatment plans are carried out in a way that is ultimately helpful for improving health and wellness. Trusted members of a person’s social network, such as family, can serve as crucial assets in promoting treatment engagement (Glick et al., 2011) and co-advocating for patients’ needs.

Physical health and wellness were also identified as important to premature mortality, which is not surprising since people with SMI are disproportionately affected by co-occurring medical conditions (Walker et al., 2015). Participants noted the roles of poor overall physical health, poor health behaviors, and obesity in particular as contributing to early mortality, with physical activity as a strategy for preventing mortality. Leveraging the power of social connectedness via peer-led integrated interventions that address comorbid physical and mental health conditions, in part through modeling new physical activities for patients with limited experience (Johnstone et al., 2009; Muralidharan et al., 2021), may be particularly useful in this domain given evidence that they can improve both physical and mental health-related quality of life (Druss, Singh, et al., 2018).

Finally, participants described the important roles of resilience and mental health in mortality among people with SMI. Mental health problems have been identified as barriers to engaging in health behaviors (Pearsall et al., 2014). People with SMI are also at increased risk for suicide, a point raised by several participants, in part due to interpersonal problems (Schmutte et al., 2021). Hopelessness, depression, and lack of motivation, also identified by our participants, are key risk factors for suicide, with technology-based interventions identified as possible suicide prevention strategies. In contrast, resilience factors identified by our participants included hope, optimism, purpose in life, and self-worth, as well as access to new opportunities. Hope, previously identified by PSSs as an important target of intervention for people with SMI (Fortuna, Ferron, et al., 2019), optimism, purpose, and self-worth play foundational roles for physical and mental health. They can promote engagement in care (and therefore improve health), while also making life worth living and health worth maintaining. Importantly, social relationships can provide the context for many of these resilience factors; for example, participants in this study referenced their children as giving them purpose in life, while also noting that having “people to talk to” during vulnerable moments can aide suicide prevention. Although the medical model of care continues to prevail, with most psychosocial interventions still emphasizing clinical outcomes over quality-of-life outcomes, “whole health” approaches to care are gaining in popularity. Ideally, both clinical and functioning/quality-of-life outcomes will be increasingly considered in treatment development, outcome research, and program monitoring among people with SMI.

There are few studies of interventions directly and primarily targeting social connectedness among people with SMI. Social skills training has been the primary intervention target to address loneliness in people with schizophrenia, yet these approaches omit social-cognitive, metacognitive, and other psychological intervention targets. Focusing on such targets among people with SMI could accelerate progress in treatment development across the continuum of care (Alegría et al., 2018). Interventions that leverage relationships to improve physical health, such as peer-led and family-assisted interventions, have been shown to be effective; mechanistic examination of these interventions can determine to what degree outcomes can be attributed to connectedness with the PSSs (Aschbrenner, Mueser, et al., 2017; Druss, Singh, et al., 2018). Additionally, improving social connectedness could have additive, synergistic, and downstream effects on other risk and protective factors, including those identified by our sample. For example, stronger relationships have been shown to be associated with stronger recovery attitudes and less internalized stigma (Cullen et al., 2017). Close others can foster adaptive coping and resilience, promote engagement in healthy activities and habits, and encourage treatment participation (Southwick et al., 2016). Utilizing close others for support may be especially important during care transitions during which people with SMI experience significant challenges (Storm et al., 2019). Thus, it may be that interventions that address social connectedness while also addressing health behaviors and resilience factors (e.g., “whole health”) may be most successful in preventing early mortality.

In general, challenges to studying and intervening on relationships exist, including poor quality measures (Hare-Duke et al., 2019), lack of standard definitions and measures of social connectedness and which qualities are most integral to which outcomes, and poor understanding of intersections with other SDOH. Additionally, there are important considerations when focusing on SMI patients’ relationships. They are not a unitary, homogenous group, and their loneliness and poor social connectedness may appear secondary to more acute health needs. In spite of these challenges, we encourage investigators to consider and test intervention strategies ranging from lighter touch self-management digital interventions, peer support, and their combination (Fortuna, Brooks, et al., 2019) to more intensive family-assisted interventions (Aschbrenner, Mueser, et al., 2017; Cohen et al., 2013). We also recommend examination of the biological impact of loneliness and interventions that target loneliness among people with SMI in order to elucidate biological mechanisms of premature mortality, such as telomere length and inflammation, which could themselves serve as novel targets of interventions (Uchino, 2006).

Limitations of the present study include our convenience sample derived from one agency within one state/one country. We asked open-ended questions and did not specifically query about many SDOH that may be relevant for understanding premature mortality in people with SMI. Importantly, the interviews were intended for a different purpose than reviewed here, and we focused analyses on the two questions most pertinent to our research questions; it is possible that content relevant to the current research questions was raised in other parts of the interview. The context of the broader interview focused on peer support may have influenced participants’ responses to these two questions. Heterogeneity of perspectives based on diagnosis was not explored. Despite these limitations, to the authors’ knowledge, this is the first study to explore the perspectives of SUs and certified PSSs on the actual causes of death of people with SMI.

In conclusion, based on current and previous findings, we recommend increased and sustained attention to promoting protective factors such as social connectedness in addition to reducing risk factors among individuals with SMI in an effort to improve health and increase longevity. Enduring chronic stress and loneliness may provide a psychobiological environment for early mortality, whereas strong relationships can provide a platform for health—both directly and indirectly by promoting engagement in treatment, adaptive coping, increasing resilience, and improved health behaviors.

Impact and Implications.

Individuals with serious mental illness (SMI) die earlier than those without. We interviewed certified peer support specialists and service users to explore potential reasons for this mortality gap. They most frequently identified social connectedness as relevant to mortality, as well as treatment, coping, physical health/wellness, and resilience/mental health. Interventions that promote social connectedness in conjunction with health behaviors may be most effective for preventing early death among people with SMI.